When a company (or other group with vested interest) conducts its own research and publishes its own analysis of the results, it is usually worth investigating more deeply. Turnitin has long been a favourite source of disingenuous disinformation (see. for instance, my posts How much plagiarism?, Guilty: how do you plead?, A second look at SEER, and Not as I do, but… ).

Now my attention turns to RefME, the reference generator (unless it is a citation generator; there may be language differences here, as discussed in Language and labels).

RefME has just published a report Survey Reveals Unique Insights to US Students’ Attitudes Towards Plagiarism on two surveys which the company carried out in recent months. It seems a prime example of how not to analyse data, how not to write a report. That’s a brutal assessment, but I think the brutality is justified. Just be sure to get in quick in case the report is edited or deleted.

I think there are (at least) five or six ways in which the report can be considered flawed. Fuller explanation follows the list:

1. the discussion of the surveys reads at times like an inadequate discussion of the surveys and at times like a press release produced by the RefME publicity bureau;
2. the report manages to confuse and conflate incorrect or inconsistently formatted references with plagiarism and/or academic misconduct;
3. the discussion grabs at different research and studies, and suggests (inter alia) that small-scale surveys can be regarded as universal truths;
4. in grabbing at those different research reports and studies, the writer misreports some and fails to do the homework, to check on the source behind the source;
5. the report, despite praising RefME for enabling correct and consistent referencing/ endnoting, manages to be incorrect, incomplete and/or inconsistent in at least 11 of its 13 references.
6. a small matter of several, many, passages which reuse so much wording from source documents that it might be felt that quotation marks are required; some readers might even class these passages as plagiarism.

This is not to denigrate the RefME software itself. I have no opinion there. Until I bought a new computer a few months ago, I found the app hung up too often to enable a valid critique of its performance as a reference (or citation) generator. Now, I find it Continue reading →